In operation management work of a system (system to be managed) requiring high reliability and high availability, an autonomous system is now being introduced. The autonomous system is a system operating according to rules such as an if-then type rule. The autonomous system assumes in advance given conditions that are likely to occur in a system to be managed, describes countermeasures (actions) as rules for responding to the assumed conditions if they happen, and stores the rules therein. When the system to be managed falls into the condition specified in the rules, the autonomous system automatically executes the countermeasure (action) specified in the rules for the system to be managed. As described above, the autonomous system is configured to immediately execute a correct countermeasure for the system to be managed.
However, many administrators are distrustful of the autonomous system which autonomously executes processing related to a critical issue such as countermeasure against a system fault out of sight of them. Such administrators are likely to hesitate to introduce the autonomous system.
Further, the autonomous system executes the abovementioned actions to the system to be managed according to rules. Therefore, in the case where the system to be managed falls into an unknown condition that is not described in the rules, the system cannot autonomously respond to it. In such a case, the condition requiring manual operation may be left untreated.
Further, it is substantially impossible to assume in advance all conditions possibly occurring in the system to be managed and set rules completely covering the assumed conditions. In addition, in order to create countermeasures having higher reliability, a rule creator needs to have high level technical knowledge and responsibility and, therefore, a large burden is imposed on the rule creator.
There have been proposed systems for managing policies (rule) (see, e.g., Patent Document 1). A policy management system for managing various kinds of policy for managing and controlling a network, which has been described in Patent Document 1, is provided with a means for setting up a plurality of policy data corresponding to each of a plurality of management target identification information corresponding to the network and a means for referring to a policy application state data to be updated and stored on the basis of the execution result of a command for setting the policy data updated and stored by the setting means in the network and selecting a suitable command for setting up the policy data in the network. With the above configuration, the policy management network disclosed in Patent Document 1 reduces a burden on an administrator involved in policy setting.
However, the system disclosed in Patent Document 1 needs to be provided with a setting means corresponding to the policy data. Thus, policy setting involving manual operation or policy setting of information such as a procedure manual cannot be handled by administrators other than those having high level knowledge and experience.
Patent Document 2 discloses a network management system. This network management system is provided with a means for detecting a fault occurring on a network, a means for storing operating indexes used for fault countermeasure, a means for taking out an operating index corresponding to a detected fault, a means for displaying the operating index, and a means for an operator to execute fault countermeasure according to the operating index, thereby enabling a network administrator to quickly take a countermeasure against the fault.
However, in the system disclosed in Patent Document 2, the operating indexes need to be previously classified according to the type of all the faults possibly occurring and, in the case where a fault that the system has not assumed occurs, fault countermeasure operation stops.    Patent Document 1: JP-A-2002-190800 (paragraphs 0035 to 0104)    Patent Document 2: JP-A-05-244156 (paragraphs 0007 to 0014)